Prenatal Exposure to Unconventional Oil and Gas Operation ......Prenatal Exposure to Unconventional Oil and Gas Operation Chemical Mixtures Altered Mammary Gland Development in Adult

R E S E A R C H A R T I C L E

Prenatal Exposure to Unconventional Oil and GasOperation Chemical Mixtures Altered MammaryGland Development in Adult Female Mice

Sarah A. Sapouckey,1 Christopher D. Kassotis,2 Susan C. Nagel,2

and Laura N. Vandenberg1

1Department of Environmental Health Sciences, School of Public Health and Health Sciences, University ofMassachusetts, Amherst, Massachusetts 01003; and 2Department of Obstetrics, Gynecology, andWomen’sHealth, University of Missouri, Columbia, Missouri 65211

Unconventional oil and gas (UOG) operations, which combine hydraulic fracturing (fracking) anddirectional drilling, involve the use of hundreds of chemicals, including many with endocrine-disrupting properties. Two previous studies examined mice exposed during early development to a23-chemical mixture of UOG compounds (UOG-MIX) commonly used or produced in the process.Both male and female offspring exposed prenatally to one or more doses of UOG-MIX displayedalterations to endocrine organ function and serum hormone concentrations. We hypothesized thatprenatal UOG-MIX exposure would similarly disrupt development of the mouse mammary gland.Female C57Bl/6 mice were exposed to ~3, ~30, ~ 300, or ~3000 mg/kg/d UOG-MIX from gestationalday 11 to birth. Although no effects were observed on themammary glands of these females beforepuberty, in early adulthood, females exposed to 300 or 3000 mg/kg/d UOG-MIX developed moredense mammary epithelial ducts; females exposed to 3 mg/kg/d UOG-MIX had an altered ratio ofapoptosis to proliferation in the mammary epithelium. Furthermore, adult females from all UOG-MIX–treated groups developed intraductal hyperplasia that resembled terminal end buds (i.e.,highly proliferative structures typically seen at puberty). These results suggest that the mammarygland is sensitive to mixtures of chemicals used in UOG production at exposure levels that areenvironmentally relevant. The effect of these findings on the long-term health of the mammarygland, including its lactational capacity and its risk of cancer, should be evaluated in future studies.(Endocrinology 159: 1277–1289, 2018)

Unconventional oil and gas (UOG) operations com-bine hydraulic fracturing (fracking) and directional

drilling. These techniques were developed to collect de-posits of oil and natural gas found in deep undergroundshale beds in low-permeability geologic formations (1).During the fracking process, a mixture of water andchemicals is pumped deep into the shale bed under highpressure, fracturing the reservoir rock, and releasingdeposits of gas and/or oil, which can then be recoveredat the surface. More than 1000 different chemicals arereportedly used during UOG operations for a range ofpurposes, including compounds that act as bactericides,stabilizers for the clay in the ground, chemicals that alterfriction and fluid viscosity, and others, although each

individual site typically uses only 12 to 24 of thesecompounds. During UOG operations, up to severalmillion gallons of water are injected per well, and amixture of injected fluids and target formation water arecollected throughout the life of the producing well.With.17million Americans living within 1mile of an oiland gas well (2), concerns have been raised about thepossibility of contamination of surface and groundwaterby the released oil and gas, the numerous inorganiccompounds that are liberated from target geologic layers(e.g., tracemetals, radioactive isotopes,minerals), and thechemicals used in well injection (3, 4).

More than 1000 chemicals have been identified inhydraulic fracturing fluids and waste water and/or have

ISSN Online 1945-7170Copyright © 2018 Endocrine SocietyReceived 27 September 2017. Accepted 19 December 2017.First Published Online 7 February 2018

Abbreviations: AR, androgen receptor; EDC, endocrine-disrupting chemical; ER,estrogen receptor; PND, postnatal day; TEB, terminal end bud; TUNEL, terminal deoxy-nucleotidyltransferase-mediated dUTP nick end labeling; UOG, unconventional oiland gas; UOG-MIX, 23-chemical mixture of unconventional oil and gas compounds.

doi: 10.1210/en.2017-00866 Endocrinology, March 2018, 159(3):1277–1289 https://academic.oup.com/endo 1277

Downloaded from https://academic.oup.com/endo/article-abstract/159/3/1277/4841948by University of Nottingham useron 08 May 2018

been reported to be used by the industry (5, 6). Many ofthese chemicals are known developmental and repro-ductive toxicants (7). Furthermore, recent evaluationsfound that .100 of these chemicals are known or sus-pected endocrine-disrupting chemicals (EDCs) (4, 8–10)(i.e., compounds that interfere with hormone action)(11). Water samples collected in drilling-dense or UOGwastewater-affected areas of the United States haveexhibited disruption of the estrogen, androgen, proges-terone, glucocorticoid, and thyroid receptors (4, 8, 12).

In 2015, Kassotis et al. (13) evaluated 24 chemicalsthat were reported by industry as commonly used and/orproduced by UOG operations to determine whether theydisplayed endocrine-disrupting properties.With cell-basedreporter gene assays, their study revealed antiandrogenic,antiestrogenic, antiprogestogenic, antithyroidogenic,and antiglucocorticogenic activities for many of thesecompounds. When evaluated as mixtures, additive and,in some cases, synergistic antagonism of these receptorswas also observed. Kassotis et al. (13) then evaluated theeffects of a 23-chemical mixture of UOG chemicals(UOG-MIX) on male mice. This mixture included the 24chemicals originally assessed, absent bisphenol A, a well-characterized EDC that has been evaluated at lengthpreviously (14, 15) but that is not directly used in UOGextraction, as reported by the industry (16). Only one ofthe chemicals included in this list (Table 1), benzene, wasevaluated in a review of 216 chemicals for carcinogeniceffects in the mammary gland, highlighting the generallack of knowledge on these chemicals (17). Male miceexposed to environmentally relevant doses of UOG-MIXduring prenatal development displayed an increasedtesticular weight before puberty and, in adulthood, de-creased sperm counts, increased serum testosteroneconcentrations, and alterations to the weight of addi-tional organs, including the heart and thymus (13).A second study revealed the effects of developmentalexposures to UOG-MIX on the female siblings (18).Exposed females had alterations to the number and de-velopmental stages of ovarian follicles measured be-fore puberty and in adulthood. The weight of severalorgans, including the uterus, ovary, and heart, wasalso affected. Furthermore, the serum concentrations ofseveral hormones were disrupted in these exposed fe-males. These results suggest the possibility that otherhormone-sensitive organs might be disrupted by expo-sure to UOG chemical mixtures.

The mouse mammary gland has proved to be an ex-cellent model to study the effects of EDCs (19–21). Thedevelopment of the mammary gland is dependent on es-trogen, progesterone, prolactin, testosterone, and growthhormone, making it an integrated biological endpoint thatis sensitive to the agonists and antagonists of different

hormone receptors (22). The estrogen receptor (ER) isexpressed in the mesenchymal compartment of the fetalgland; however, its expression shifts to the epithelialcompartment in the adult gland (23). Although ERaknockout mice have mammary glands that are in-distinguishable from those of wild-type controls beforepuberty, they are visibly stunted compared with controlsonce puberty begins, highlighting the importance of es-trogen for growth of the gland (24, 25). The androgenreceptor (AR) is also expressed in the mesenchyme of thefetal mammary gland, and the testosterone produced bythe testes in male fetuses causes the mesenchyme to con-dense around the epithelium, detaching the epitheliumfrom the skin. Thus, the male mice will typically not havenipples (26, 27). Antiandrogenic compounds can lead tonipple retention in exposed males (28). Testosterone islikely to have a physiological role during postnatal de-velopment of the female gland because the mammaryglands of female AR knockout mice show impaired ductalgrowth in postnatal life, indicating a role in ductal elon-gation (29). A wide range of EDCs, including severalpharmaceutical compounds, naturally occurring EDCs,and industrial compounds with varied modes of action,have been demonstrated to alter the development of the

Table 1. Chemicals in UOG Mixture and HormoneReceptor Antagonist Activity, as Listed in (13)

Chemical Name

Receptor AntagonistActivity

ER AR PR TR GR

1,2,4-Trimethylbenzene2-(2-Methoxyethoxy) ethanol X X X2-Ethylhexanol X X X X2-Methyl-4-isothiazolin-3-one X X XAcrylamide XBenzene X XBronopol X X X XCumene X X XDiethanolamine X X XEthoxylated nonylphenol X X X X XEthoxylated octylphenol X X X X XEthylbenzene X XEthylene glycol X X X X XEthylene glycol monobutyl ether X X XNaphthalene X X X X XN,N-dimethylformamide X X XPhenol X X XPropylene glycol XSodium tetraborate decahydrate X XStyrene X X XToluene X XTriethylene glycol X X

X indicates the presence of receptor antagonist activity as measured viatransiently transfected reporter gene assays in human cells.

Abbreviations: GR, growth receptor; PR, progesterone receptor; TR,thyroid receptor.

1278 Sapouckey et al UOG Chemical Mixtures Disrupt Mouse Mammary Gland Endocrinology, March 2018, 159(3):1277–1289


mammary gland, with effects that typically manifest atpuberty and in adulthood, when endogenous hormonesinduce its growth (20, 30). The characteristics that arecommonly evaluated to determine the level of developmentand the disruption of development by EDCs include anumber of branching points, extension of the ductal epi-thelium into the stroma, the number and size of termi-nal end buds (TEBs) (e.g., highly proliferative epithelialstructures present during puberty), and the presence ofalveolar buds and lobuloalveolar units (e.g., structures thatwill produce milk in lactating females), among others (31).

Based on the hormone receptor antagonism of the 23-chemical mixture and the effects this mixture induced inthe endpoints relevant to the hypothalamic-pituitary-gonadal axis in exposed mice, we hypothesized thatprenatal exposure to the UOG-MIX would disrupt de-velopment of the mouse mammary gland. We evaluatedthe mammary glands in female mice after gestationalexposure to one of four doses of the UOG-MIX. Con-sistent with our hypothesis, we characterized the sub-stantial effects of this mixture on mammary glandmorphology in adulthood and documented the presenceof hyperplastic lesions.

Materials and Methods

Animal husbandry and chemical administrationC57BL/6J mice were housed in sterile polysulfone cages

under temperature- and light-controlled (12-hour light, 12-hourdark) conditions in a barrier animal facility. The mice were fedLabDiet 5053 and provided acidified water ad libitum fromglass bottles. Ten-week-old mice were mated, and the day ofvaginal plug formation was denoted as gestational day 0. Ongestational day 11, the dams were randomized to the treat-ment groups and provided with experimental treatment in theirdrinking water. The test concentrations included a 0.2% etha-nol vehicle; flutamide, a known antiandrogenic pharmaceuticalagent (mechanistic antiandrogen control, 167 mg/mL); and fourconcentrations of the 23-chemicalmixture (UOG-MIX),with eachindividual chemical present at 0.01, 0.10, 1.0, and 10 mg/mL. Thecomposition of the chemical mixture and hormone recep-tor antagonist activity are listed in Table 1. Water intake wasmonitored byweighing the drinkingwater bottle every day of theexperiment. Based on intake, the chemical exposure was esti-mated at 3, 30, 300, or 3000 mg/kg body weight/d for themixtures. These treatment groups are referred to asMIX-3,MIX-30, MIX-300, and MIX-3000, respectively. Intake of the anti-androgenic control was estimated at 50 mg flutamide/kg bodyweight/d. Water intake did not differ in experimental groupsrelative to the vehicle control (data not shown). Experimentaltreatment was provided until birth; the dams were then revertedto standard acidified water when the pups were first detected.Litters with ,2 males and females were removed from the an-alyses owing to concerns of gestational hormone exposure.Litters with .2 males and females were left unaltered, becauseculling within litters has been shown to alter the feeding, be-havior, and physiology of the remaining pups (32).

All experimental procedures were performed according to anapproved University of Missouri Animal Care and Use Com-mittee protocol and were in accordance with the NationalResearch Council’s Guide for the Care and Use of LaboratoryAnimals.

Tissue collectionAll mice were euthanized by carbon dioxide asphyxiation

and cardiac puncture, and the mammary tissues were excised.One randomly selected female pup from each litter underwentnecropsy on either postnatal day (PND) 21 or PND85. At bothages, one fourth of the inguinal mammary gland was wholemounted on a glass slide, fixed in neutral-buffered formalinovernight, and processed for carmine staining using proto-cols described previously (33). Whole mounts were preserved inK-pax sealed bags with methyl salicylate. The contralateralfourth inguinal mammary gland was formalin-fixed for 24hours, washed in phosphate-buffered saline, and stored in 70%ethanol for histological evaluation.

Morphological analysis of whole mountmammary glands

Whole mount specimens were imaged using a Zeiss dissectingmicroscope with AxioCam HRc digital camera. For analyses ofthe prepubertal mammary gland (PND21), imageswere capturedof the full ductal tree and its position relative to the central lymphnode. The following measurements were taken using methodsdeveloped previously (33): total ductal area, measured byquantifying the area subtended by the ducts; ductal extension,quantified as the furthest growthof the ductal treemeasured fromthe center of the lymph node; and the number of branchingpoints, counted throughout the entire gland. Nomice had visibleTEBs; thus, these structures were not quantified.

For analyses of the adult gland, two photographswere taken,one of the entire mammary gland (at 33 magnification) andone anterior to the central lymph node (at 313.5 magnifica-tion); the former was used to evaluate the presence of TEB-likestructures and the latter for unbiased stereological evaluationsusing ZEN imaging software (Zeiss). In brief, to quantify thevolume fraction of epithelial structures, a 130-point grid wassuperimposed over each anterior photograph. The structurepresent on each crosshair was counted; the individual epithe-lial structures that were evaluated included ducts (when thecrosshair hit the middle of a duct), terminal ducts/terminal ends(when the crosshair hit a blunt end of a duct), and alveolar buds(34). The volume fraction of each type of structure was cal-culated by counting the number of crosshairs that hit eachstructure divided by the total number of crosshairs hitting themammary tissue (typically, 130). Volume fraction of all epi-theliumwas calculated by summing all epithelial structures (i.e.,ducts, terminal ends, and alveolar buds).

Excision of mammary tissue from whole mountsFor whole mounts that displayed unusual or TEB-like

structures, small areas of the gland, including these abnormalstructures, were excised using a scalpel and dissecting micro-scope. The remainder of the whole mount was then rebaggedusing methyl salicylate. The excised parts of the whole mountswere washed, processed through a series of alcohols, embeddedin paraffin, and sectioned using the methods described in the

doi: 10.1210/en.2017-00866 https://academic.oup.com/endo 1279


following sections. Additional areas from the whole mountspecimenswith a normal appearancewere also excised for use ascontrol tissues. All excised tissues were coded such that addi-tional analyses could be conducted by experimenters who wereunaware of their origin (e.g., unusual vs normal).

Histological evaluationOne mammary gland from each sample was processed

through a series of dehydrating alcohol washes and embedded inparaffin under vacuum; 5-mm sections were produced using arotary microtome and placed on Superfrost positively chargedslides (Fisher Scientific). This process was also used to evalu-ate excised tissue from whole mount mammary glands. For thehistological evaluations, the sections were deparaffinized inxylene and rehydrated in a series of alcohol washes. They werethen stained usingHarris hematoxylin and eosin (Fisher Scientific),dehydrated through an alcohol series, washed in xylene, andmounted using a permanent mounting medium (Fisher Scientific).The slides were examined using a Zeiss Observer Z1 inverted lightmicroscope at340 magnification. Images were captured using anAxioCam HRc digital camera and evaluated with ZEN imagingsoftware (Zeiss) for the presence of hyperplastic ducts.

ImmunohistochemistryImmunohistochemical analysis for ERa and Ki67, a marker

of proliferation, was performed as previously described (34).Primary antibodies were used at 1:1000 (ERa; catalog no. 06-935; Millipore; Ki67; catalog no. 9106-S1; Thermo FisherScientific; Table 2). For feasibility, the samples were examinedin the control,MIX-3, andMIX-3000 groups only; these groupswere selected to evaluate a wide range of doses. The sampleswere visualized using a Zeiss Observer Z1 inverted light mi-croscope at 340 magnification, and the images were capturedusing an AxioCam HRc digital camera. The images were an-alyzed using ZEN imaging software (Zeiss). For each sample,two or three fields of view were selected arbitrarily (dependingon the size of the epithelial ducts) and imaged at 340; theexpression of each marker was quantified in all ducts withinthese images. At least 200 epithelial cells were assessed for eachantigen; each cell was counted as either positively expressing themarker of interest (brown owing to diaminobenzidine, thecolorimetric reaction used to visualize immunohistochemicalreactions), or no expression (blue; hematoxylin counterstain).

Terminal deoxynucleotidyltransferase-mediateddUTP nick end labeling assay

The Trevigen TACS 2 TdT-DAB in situ apoptosis detectionkit was used for detection of apoptotic cells in the mammarytissue sections. All samples were counterstained with Harris

hematoxylin, dehydrated, mountedwith a permanent mountingmedium, and imaged with a Zeiss Observer Z1 inverted lightmicroscope at 340 magnification. Quantification of terminaldeoxynucleotidyltransferase-mediated dUTP nick end labeling(TUNEL) was completed in two or three fields of view selectedarbitrarily; all ducts within these images were evaluated. At least200 epithelial cells were assessed; each cell was counted as eitherTUNEL-positive (brown owing to diaminobenzidine, the col-orimetric reaction used to visualize the TUNEL reactions) or noexpression (blue; hematoxylin counterstain).

Statistical analysisThe SPSS statistical software package, version 22 (IBM

Corp.), was used for all statistical analyses. Analysis of variancefollowed by Fisher post hoc tests, was used to assess for dif-ferences between the control and UOG-MIX treatment groupsfor each age (PND21 and PND85). Independent samples t testswere used to compare the control and flutamide-treated groups(35). A x2 test was performed to compare the incidence ofhyperplasia in the mammary gland epithelium of control andUOG-MIX–exposed mice. To account for litter effects, only 1mouse was selected from each litter for each age evaluated. Forall statistical tests, the results were considered statistically sig-nificant at P , 0.05. All results are presented as the mean 6standard error of the mean and were collected and analyzed byexperimenters who were unaware of the treatment.

Results

Prepubertal mammary gland morphology was notaltered by developmental exposure to UOG-MIX

The mice were exposed to vehicle or one of four dosesof a 23-chemical mixture (UOG-MIX) during gestation.To determine the effects of the UOG chemicals on thefemale prepubertal mammary gland, morphologicalevaluations were first conducted at PND21. TEBs werenot present in any glands from any treatment group,consistent with the prepubertal stage of mammary glanddevelopment (Fig. 1A; data not shown). Morphometricevaluations revealed an inverse association betweentreatment and total ductal area (e.g., smaller epithelialtrees in UOG-MIX–treated females), although no sta-tistically significant differences were found in this growthparameter between controls and the UOG-MIX treat-ments (Fig. 1B). Flutamide-treated females had signifi-cantly smaller ductal trees compared with the controls

Table 2. Antibody Table

Peptide/ ProteinTarget

AntigenSequence(if Known)

Name ofAntibody

Manufacturer; CatalogNo.

Species Raised in;Monoclonal or

Polyclonal RRIDDilutionUsed

ERa Anti-ERa (C1355) Millipore; 06-935 Rabbit; polyclonal AB_310305 1:1000Ki67 Ki67 Thermo Fisher Scientific;

RM-9106-S1Rabbit; monoclonal AB_149792 1:1000

Secondary Biotinylated goatanti-rabbit IgG

Abcam; ab64256 Goat; polyclonal AB_2661852 Ready to use(5 mg/mL)

Abbreviations: IgG, immunoglobulin G; RRID, Research Resource Identifier.



(P , 0.05; independent samples t test; Fig. 1B). Ductalextension and the total number of branching points werenot different in any treatment group, including theflutamide-treated females (Table 3). Collectively, thesedata suggest that prenatal exposure to the 23-chemicalmixture did not alter the morphological features of thefemale mammary glands before puberty.

Adult mammary gland morphology was altered bydevelopmental exposure to UOG chemical mixtures

Increased volume of ducts and epithelialcompartment

Unbiased stereological methods revealed treatment-related effects on mammary gland morphology in earlyadulthood (PND85; Fig. 2A). Females exposed to MIX-300 had significantly more ducts comparedwith vehicle-treated controls (P , 0.05, Fisher post hoc test; Fig. 2Aand 2B), and a trend for an increase was also seen in

females exposed to MIX-30 and MIX-3000 (P , 0.1,Fisher post hoc test; Fig. 2A and 2B). The volumefraction of total mammary epithelium was significantlyincreased in females exposed to MIX-300 and MIX-3000 (P , 0.05, Fisher post hoc test; Fig. 2A and2C). Neither of these parameters were significantly al-tered by flutamide treatment. Alveolar buds were notobserved in females of any treatment group (Fig. 2A;data not shown).

Increased proliferation/apoptosis ratioThe growth of the mammary epithelium is dependent

on a balance of proliferation (to extend ductal structuresinto the mammary fat pad) and apoptosis (to producehollow ducts capable of transporting milk) (36). Toquantify apoptosis in the epithelium of the adult fe-male mammary glands, TUNEL staining was used tocompare the controls,MIX-3,MIX-3000, and flutamide-treated mice. No statistically significant changes inTUNEL incorporation were seen in MIX-3 and MIX-3000 females (Fig. 3A and 3B). A borderline statisticallysignificant decrease was seen in TUNEL-positive cells inthe flutamide-treated group (P = 0.052, independentsamples t test; Fig. 3B).

We next evaluated proliferation in the mammaryepithelium using antibodies for Ki67, a marker of pro-liferation. Although the proliferation levels were low, asexpected for adult mammary glands, we observed sta-tistically significant effects of the 23-chemical mixture onthe number of cells expressing Ki67 in the MIX-3 group(P, 0.05; Fisher post hoc test; Fig. 3A and 3C). Females

Figure 1. No substantial effects from the developmental exposure to fracking mixtures were observed on the morphology of the prepubertalmammary gland. (A) Examples of whole mount mammary glands from vehicle, MIX-3000, and flutamide (FLUT)-treated female mice. Scalebar = 0.5 mm. (B) Quantification of ductal area suggested an inverse relationship between UOG-MIX treatment and ductal area, although thisdifference was not statistically significant. The flutamide-treated female mice had significantly smaller ductal trees. ^P , 0.05, independentsamples t test, comparing control and flutamide-treated female mice. LN, lymph node.

Table 3. UOG-MIX Treatment Did not Alter GrowthParameters in Prepubertal Mammary Gland

VariableDuctal

Extension (mm)aNumber of

Branching Points

Control (n = 6) 213.64 6 1.07 25.0 6 2.2MIX-3 (n = 8) 214.19 6 0.96 24.9 6 2.1MIX-30 (n = 5) 212.96 6 0.43 20.8 6 2.3MIX-300 (n = 6) 215.74 6 1.46 25.7 6 2.5MIX-3000 (n = 6) 213.88 6 0.34 26.2 6 2.4Flutamide (n = 9) 214.03 6 0.84 20.9 6 1.4

aNegative values for ductal extension indicate ductal trees that have notyet grown past the central lymph node.



from the MIX-3 group had 427% more Ki67-positivecells compared with the controls. The MIX-3000 treat-ment group had 54% more Ki67-positive cells comparedwith the controls, although these differences were notstatistically significant. Ki67 expression was not alteredby the use of flutamide (Fig. 3C).

We evaluated the ratio of proliferation/apoptosis inthe control, MIX-3, MIX-3000, and flutamide-treatedfemales. We found a striking 397% increase in theproliferation/apoptosis ratio in the MIX-3 treatmentgroup (P, 0.05; Fisher post hoc test) and nonsignificantincreases in theMIX-3000 group (131%) compared withthe controls (Fig. 3D). No effect on the proliferation/apoptosis ratio was observed in the flutamide-treatedfemales (Fig. 3D).

ERa expression tended to be associated withdevelopmental UOG-MIX treatment

EDCs have been shown to not only bind to hormonereceptors, but also to alter expression of hormone re-ceptors in a dose-, age-, and tissue-specific manner (37).We next asked whether developmental exposure tofracking chemicals would alter expression of ERa in themammary epithelium. We observed a general increasein ERa expression with an increasing UOG-MIX dose,although these differences were not statistically sig-nificant (Fig. 3A and 3E). Flutamide also did not af-fect the percentage of epithelial cells expressing ERa(Fig. 3E).

TEB-like intraductal hyperplasia in mammary glandsafter developmental mix treatment

A striking observationmade during themorphologicalassessment of the adult (PND85) mammary glands wasthe appearance of structures that resembled TEBs (Fig. 4).These TEB-like structures were not observed in thecontrol females (Table 4). The unusual structures wereexcised from whole mount mammary glands and furthercharacterized using histological and immunohistochem-istry tools (Fig. 5). Hematoxylin and eosin stainingrevealed ducts with excessive layers of epithelial cells,consistent with intraductal hyperplasia (Fig. 5A). TheseTEB-like lesions were highly proliferative: .40% ofepithelial cells in lesions were Ki67-positive comparedwith ,4% of epithelial cells in normal ducts collectedfrom the same mice or ducts excised from control females(Fig. 5A and 5B). ERa expression was also greater inmany, although not all, of the excised lesions (Fig. 5A and5C). Collectively, these results suggest that the retainedTEB-like structures are highly proliferative intraductalhyperplasias that are likely to be estrogen-responsive.

Discussion

We examined the effects of exposure to a mixture of 23chemicals used in unconventional oil and gas extraction,most of which were previously shown to exhibit antag-onistic properties on one or more hormone receptors[(13) and Table 1], including the ER, progesterone

Figure 2. Prenatal UOG-MIX treatment induced increased epithelial density in female mammary glands at adulthood. (A) Example of wholemount mammary glands (magnification 313) from control, MIX-300, and MIX-3000 treatment groups on PND85. Vehicle-treated females hadthe least dense mammary epithelium. Note that alveolar buds were not observed in any treatment group. (B) Volume fraction of ducts and (C)volume fraction of all epithelium was increased in UOG-MIX–treated groups. *P , 0.05, Fisher post hoc test; aP , 0.1, Fisher post hoc test.



receptor, and AR. Also, 21, 20, and 11 of these chemicalswere previously demonstrated to antagonize human ER,AR, and PR, respectively, in a reporter gene assay inhuman endometrial cells (summarized in Table 1). To thebest of our knowledge, we have shown for the first time

that the mouse mammary gland is sensitive to develop-mental UOG-MIX exposure, with dose-specific effects ontissue morphology (after exposure to MIX-300 andMIX-3000), cell proliferation (after exposure to MIX-3),and the induction of unique intraductal hyperplasias

Figure 3. Apoptosis, proliferation, and proliferation/apoptosis ratios were altered in UOG-MIX– and flutamide-treated females. (A) Examples ofTUNEL and immunohistochemistry testing for Ki67 and ERa in mammary gland sections from control and MIX-3–, MIX-3000–, and flutamide-treated females. Positive cells indicated by red arrows. Scale bar = 20 mm. (B) Quantification of TUNEL incorporation in epithelium from vehicle-,MIX-3–, MIX-3000–, and flutamide-treated females revealed a dose-dependent decrease in the percentage of epithelial cells undergoingapoptosis. (C) Quantification of Ki67, a marker of proliferation, in epithelial cells. (D) Substantial alterations to the ratio of cell proliferation andcell death were observed in both MIX-3 and MIX-3000 treatment groups and the flutamide-treated group. (E) Quantification of the percentage ofmammary epithelial cells expressing ERa. *P , 0.05, Fisher post hoc test; dP , 0.1, independent samples t test, comparing control andflutamide-treated females.



(observed in all MIX groups). The effects observed in thepresent study occurred at low doses; the two lowest dosegroups (MIX-3 and MIX-30) are equivalent to theconcentrations measured in drinking water in regionsexperiencing drilling and the highest dose group (MIX-3000) is equivalent to the concentrations of many UOG-MIX components measured in industry wastewater (4,13, 18, 38). In addition, the concentrations for several ofthe 23 chemicals in the UOG-MIX used in the presentstudy have not yet been determined in either drinkingwater or wastewater.

Themammary gland is a hormone-sensitive organ thatis responsive to multiple endocrine inputs during earlydevelopment. Testosterone has a unique role in estab-lishing the sexually dimorphic development of the mousemammary gland (26, 27) but is not thought to play a rolein the female or postnatally. AR expression remains highin the mammary stroma until birth (26, 39), suggestingthat UOG-MIX exposure could affect mammary devel-opment via actions at the AR. Although some endpointswere similarly affected (e.g., retention of TEBs), othereffects of UOG-MIX exposure we observed in the presentstudy were distinct from the effects of flutamide, sug-gesting that the UOGmixturemight not working throughan antiandrogenic mechanism as we originally hypoth-esized. To date, few studies have evaluated the effects ofprenatal exposures to antiandrogenic chemicals on thefemale mammary gland; most studies of antiandrogenshave examined their effects on male offspring, includingnipple retention and other disruptions tomammary glandmorphology [reviewed in (40–42)]. These studies sug-gested that evaluations of mammary glands in male mice

Figure 4. TEB-like structures observed in whole mount mammary glands from prenatally UOG-MIX–treated females on PND85. (A) Micrograph ofwhole mount mammary gland from a control mouse with typical terminal ducts. These blunt ends (arrowheads) are the normal structuresobserved at the ends of ducts. (B) These examples illustrate the TEB-like structures excised from whole mounts from UOG-MIX– and flutamide-treated females. Arrows indicate abnormal structures. Treatment groups for the individual female mice are indicated. LN, lymph node.

Table 4. Presence of TEB-Like Structures in Controlor UOG-MIX–Treated Females

Variable TEB-Like Structures, %

Control 0 (n = 9)MIX-3 29 (n = 7)MIX-30 17 (n = 6)MIX-300 25 (n = 8)MIX-3000 40 (n = 5)Flutamide 22 (n = 9)



exposed to UOG mixtures are a priority. A bettermechanistic understanding of the effects of AR antago-nists on the female mammary gland, including a widerrange of exposure to flutamide, is also needed.

Previous studies of other EDCs have shown few effectson the female gland before puberty (33, 43), suggestingthat the most obvious effects of EDCs manifest visibly

only after the onset of ovarian hormone production. Insupport of this, we observed striking effects of develop-mental exposure to UOG chemical mixtures in the adultfemale mammary gland (at PND85). Not only weremammary glands more developed in the UOG-MIX–

treated adult females, as indicated by the increased vol-ume fraction of epithelium (Fig. 2), but UOG-MIX

Figure 5. Histological and immunohistochemical evaluation of TEB-like structures excised from UOG-MIX–treated females. (A) Representativelesions from three MIX groups (lesion 1, MIX-30; lesion 2, MIX-3; lesion 3, MIX-300) were evaluated using hematoxylin and eosin (H&E) stainingand immunohistochemistry for Ki67 (a marker of proliferation) and ERa. (B) Quantification of Ki67 in excised TEB-like structures compared withKi67 expression in other regions of the same glands. (C) Quantification of ERa in excised TEB-like structures compared with ERa expression inother regions of the same glands.



treatment also altered the ratios of proliferation andapoptosis (Fig. 3), cell parameters important for dictatingthe growth and function of the mammary gland (33, 44,45). Normal growth of the mammary epithelium is de-pendent on a balance of proliferation (to extend ductalstructures into the mammary fat pad) and apoptosis(to produce hollow ducts capable of transporting milk)(36); thus, disruptions to these cellular features couldpredispose animals to mammary gland diseases (e.g.,cancer) or abnormal function (e.g., disruptions to lac-tation). Additional studies are needed to evaluate theseoutcomes in UOG-MIX–treated females. The substantialalterations to the apoptosis/proliferation ratio suggestthat the mammary glands from UOG-MIX–treated fe-males might continue to manifest complex hyperplasticand preneoplastic lesions in later adulthood. Additionalstudies are needed to evaluate this possibility.

We were also surprised to see TEB-like hyperplasticlesions in the mammary glands collected from mice in theUOG-MIX–treatment groups, as well as in the flutamide-treated mice (Fig. 4). Other EDCs have also been shownto induce development of intraductal hyperplasias, al-though these typically will have a “beaded duct,” ratherthan a TEB-like, appearance (34, 46, 47). TEBs are acharacteristic structure of glands undergoing puberty (45,48); these highly proliferative structures drive the growthof the mammary epithelium into the surrounding fat pad.Once the epithelial tree is fully formed, the TEB structuresrecede and are not seen in adult glands (19, 49). Onepossibility is that UOG-MIX exposure during early lifedelays the appearance of TEBs, and thus their presence inthe glands of adult mice is indicative of a shift in thetiming of puberty. Although the timing of vaginalopening and the age of first vaginal estrus were not af-fected by UOG-MIX exposure (18), the timing of pu-bertal growth in the mammary gland involves distinctevents (50). Alternatively, it is possible that the timing ofmammary puberty is unaffected by UOG-MIX exposureand that the retention of TEBs is indicative of failure toprogress to blunt ductal ends. Thus, the presence of TEBscould be interpreted as diminished development of themammary gland in UOG-MIX–treated groups, in con-trast to the effects of these chemical mixtures on epithelialdensity, which are more consistent with advanced de-velopment of the gland (Fig. 2). EDCs have previouslybeen shown to produce competing effects on growthparameters in the mammary gland, including some thatadvance one aspect of development and seem to delayother developmental landmarks (51). Studies of bisphe-nol A, for example, have shown that developmentalexposure can both decrease ductal extension (growth)and increase the size of TEBs (33, 44, 52–54). Theseresults are consistent with the two competing roles that

hormones such as estrogen can have in the developinggland, including its ability to promote proliferation ofsome mammary epithelial cells and inducing apoptosis ofother mammary epithelial cells inside the duct, allowingthe lumen to form (55, 56). TEBs are one of the siteswhere cancers are thought to arise; thus, delays in TEBrecession could increase the gland’s sensitivity to car-cinogens (51, 57). Future studies are needed to determinewhether these TEBs are retained into later adulthoodand whether prenatal UOG-MIX exposure increases thesensitivity of the gland to carcinogens.

The mammary gland provides an in vivo tool toevaluate the effects of EDCs with different modes ofaction, including chemicals administered as mixtures. Todate, in vivo studies of EDC chemical mixtures have beenlimited, with even fewer studies examining mammals[examples include (58–60)]. The results from in vitrostudies have suggested that compounds with a similarmode of action can have additive effects, including somexenoestrogen mixtures, which have been described asproducing “something from nothing” (61–63). In thepresent study, the assessment of the mixture of 23chemicals in human endometrial cells demonstratedsynergistic responses for ER and thyroid receptor an-tagonism and less than additive effects for AR and glu-cocorticoid receptor antagonism (13). Other groups havefound that the toxic effects of pesticides are compoundedwhen examined as formulations compared with studyingonly the so-called active ingredient (64), suggesting thatnovel insights can be gained from examining chemicalmixtures that would not be detected, or even anticipated,from examining the mixture’s components individually.Concerns have been raised that any effects observed afterexposure to mixtures will be difficult to evaluate mech-anistically because they cannot be attributed to any singlecomponent of the chemical mixture unless all individualcomponents are also evaluated. However, it is worthnoting that humans are exposed to chemical mixtures,rather than single compounds; thus, the study of mix-tures in laboratory animals might provide better un-derstanding of the human condition (65).

To date, .1000 different chemicals have been re-ported for use during unconventional oil and gas ex-traction (5). Previous studies have shown that prenatalexposure to this same 23-chemical mixture induced ad-verse health outcomes in the male offspring (the brothersof the female mice examined in the present study) (13).Furthermore, other endocrine organs were affected inthe female mice examined in the present study, in additionto alterations to serum hormone concentrations, includ-ing decreased levels of the pituitary hormones luteinizinghormone, prolactin, and follicle-stimulating hormone,among others (18). Decreased pituitary hormone



concentrations could influence mammary gland health,including the development of the gland during preg-nancy, an endpoint that deserves future attention.

Determining whether mixtures of fracking chemicalsaffect human populations is an important goal, especiallyas the number of drilling sites continues to increase (66).A recent systematic review evaluating the strength of thedata for the association between conventional oil and gasand UOG operations and human reproductive outcomesfound moderate evidence for an increased risk of pretermbirth, miscarriage, birth defects, decreased semen quality,and prostate cancer (67). The evidence for an associationbetween UOG operations and breast cancer, or otherdiseases of the breast, remains inadequate. The resultsfrom our study suggest that longitudinal studies evalu-ating women exposed to UOG chemical mixtures duringearly life are needed to address this data gap.

Conclusion

Future studies are needed to evaluate the many additionalchemicals used in, and produced by, UOG processes tobetter quantify the concentrations of these and othercontaminants in environmental samples and to assess theeffects of exposure during other sensitive windows ofdevelopment, including pregnancy and lactation, pu-berty, and the aging female. Future studies should alsoevaluate whether developmental UOG-MIX treatmentscan sensitize animals to hormones or carcinogens, aswould be expected in females with retained TEBs. Formechanistic insights, additional examination of someof the individual components in the fracking chemicalmixture might be warranted.

Acknowledgments

The authors thank the other members of the Vandenberg andNagel Laboratories for feedback on our study.

Financial Support: This work was supported by the Na-tional Institute of Environmental Health Sciences of the Na-tional Institutes of Health (Award K22-ES025811 to L.N.V.and Grants R21-ES026395 and R01-ES021394-04S1 toS.C.N.). The content of this manuscript is solely the re-sponsibility of the authors and does not necessarily represent theofficial views of the National Institutes of Health.

Correspondence: Susan C. Nagel, PhD, Department ofObstetrics, Gynecology and Women’s Health, University ofMissouri, DC051.00, One Hospital Drive, Columbia, Missouri65211. E-mail: [email protected]; or Laura N.Vandenberg, PhD, Department of Environmental Health Sci-ences, School of Public Health and Health Sciences, Universityof Massachusetts, 171A Goessmann, 686 North PleasantStreet, Amherst, Massachusetts 01003. E-mail: [email protected].

Disclosure Summary: L.N.V. has received travel reim-bursement from universities, governments, nongovernmentalorganizations, and industry to speak about endocrine-disrupting chemicals. The remaining authors have nothing todisclose.

References

1. Wilson DG. Energy supplies and future engines for land, sea, andair. J Air Waste Manag Assoc. 2012;62(6):607–624.

2. Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, AbadJD. Impact of shale gas development on regional water quality.Science. 2013;340(6134):1235009.

3. Vengosh A, Jackson RB, Warner N, Darrah TH, Kondash A. Acritical review of the risks to water resources from unconventionalshale gas development and hydraulic fracturing in the United States.Environ Sci Technol. 2014;48(15):8334–8348.

4. Kassotis CD, Iwanowicz LR, Akob DM, Cozzarelli IM, MumfordAC, Orem WH, Nagel SC. Endocrine disrupting activities of sur-face water associated with a West Virginia oil and gas indus-try wastewater disposal site. Sci Total Environ. 2016;557-558:901–910.

5. Elliott EG, Ettinger AS, Leaderer BP, Bracken MB, Deziel NC. Asystematic evaluation of chemicals in hydraulic-fracturing fluidsand wastewater for reproductive and developmental toxicity.J Expo Sci Environ Epidemiol. 2017;27(1):90–99.

6. Wylie S, Schultz K, Thomas D, Kassotis C, Nagel S. Inspiringcollaboration: the legacy of Theo Colborn’s transdisciplinary re-search on fracking [published online ahead of print September 13,2016]. New Solut. doi:1048291116666037.

7. Webb E, Bushkin-Bedient S, Cheng A, Kassotis CD, Balise V, NagelSC.Developmental and reproductive effects of chemicals associatedwith unconventional oil and natural gas operations. Rev EnvironHealth. 2014;29(4):307–318.

8. Kassotis CD, Tillitt DE, Davis JW, Hormann AM, Nagel SC. Es-trogen and androgen receptor activities of hydraulic fracturingchemicals and surface and ground water in a drilling-dense region.Endocrinology. 2014;155(3):897–907.

9. Bolden AL, Kwiatkowski CF, Colborn T. New look at BTEX:are ambient levels a problem? Environ Sci Technol. 2015;49(9):5261–5276.

10. Colborn T, Kwiatkowski C, Schultz K, Bachran M. Natural gasoperations from a public health perspective.HumEcol Risk Assess.2011;17(5):1039–1056.

11. Zoeller RT, Brown TR, Doan LL, Gore AC, Skakkebaek NE, SotoAM, Woodruff TJ, Vom Saal FS. Endocrine-disrupting chemicalsand public health protection: a statement of principles from TheEndocrine Society. Endocrinology. 2012;153(9):4097–4110.

12. Cozzarelli IM, Skalak KJ, Kent DB, Engle MA, Benthem A,Mumford AC, Haase K, Farag A, Harper D, Nagel SC, IwanowiczLR, Orem WH, Akob DM, Jaeschke JB, Galloway J, Kohler M,Stoliker DL, Jolly GD. Environmental signatures and effects of anoil and gas wastewater spill in the Williston Basin, North Dakota.Sci Total Environ. 2017;579:1781–1793.

13. Kassotis CD,KlempKC,VuDC, LinCH,MengCX,Besch-WillifordCL, Pinatti L, Zoeller RT, Drobnis EZ, Balise VD, Isiguzo CJ,WilliamsMA, Tillitt DE, Nagel SC. Endocrine-disrupting activity ofhydraulic fracturing chemicals and adverse health outcomes afterprenatal exposure in male mice. Endocrinology. 2015;156(12):4458–4473.

14. Vandenberg LN, Ehrlich S, Belcher SM, Ben-Jonathan N, DolinoyDC, Hugo ER, Hunt PA, Newbold RR, Rubin BS, Saili KS, SotoAM, Wang HS, Vom Saal FS. Low dose effects of bisphenol A: anintegrated review of in vitro, laboratory animal and epidemiologystudies. Endocr Disruptors (Austin). 2013;1:e25078.

15. Rochester JR. Bisphenol A and human health: a review of theliterature. Reprod Toxicol. 2013;42:132–155.



16. Environmental Protection Agency. Assessment of the potential im-pacts of hydraulic fracturing for oil and gas on drinking water re-sources. Available at: http://www.epa.gov/hfstudy2015. Accessed 7September 2017.

17. Rudel RA, Attfield KR, Schifano JN, Brody JG. Chemicals causingmammary gland tumors in animals signal new directions for epi-demiology, chemicals testing, and risk assessment for breast cancerprevention. Cancer. 2007;109(12 suppl)2635–2666.

18. Kassotis CD, Bromfield JJ, Klemp KC,Meng CX,Wolfe A, ZoellerRT, Balise VD, Isiguzo CJ, Tillitt DE, Nagel SC. Adverse re-productive and developmental health outcomes following prenatalexposure to a hydraulic fracturing chemical mixture in femaleC57Bl/6 mice. Endocrinology. 2016;157(9):3469–3481.

19. Rudel RA, Fenton SE, Ackerman JM, Euling SY, Makris SL. En-vironmental exposures and mammary gland development: state ofthe science, public health implications, and research recommen-dations. Environ Health Perspect. 2011;119(8):1053–1061.

20. Fenton SE. Themammary gland: a tissue sensitive to environmentalexposures. Rev Environ Health. 2009;24(4):319–325.

21. Soto AM, Vandenberg LN, Maffini MV, Sonnenschein C. Doesbreast cancer start in the womb? Basic Clin Pharmacol Toxicol.2008;102(2):125–133.

22. Brisken C, Ataca D. Endocrine hormones and local signals duringthe development of the mouse mammary gland. Wiley InterdiscipRev Dev Biol. 2015;4(3):181–195.

23. Saji S, Jensen EV,Nilsson S, Rylander T,WarnerM,Gustafsson JA.Estrogen receptors alpha and beta in the rodent mammary gland.Proc Natl Acad Sci USA. 2000;97(1):337–342.

24. Bocchinfuso WP, Korach KS. Mammary gland development andtumorigenesis in estrogen receptor knockout mice. J MammaryGland Biol Neoplasia. 1997;2(4):323–334.

25. Mallepell S, Krust A, Chambon P, Brisken C. Paracrine signalingthrough the epithelial estrogen receptor alpha is required forproliferation and morphogenesis in the mammary gland. Proc NatlAcad Sci USA. 2006;103(7):2196–2201.

26. Kratochwil K. Development and loss of androgen responsiveness inthe embryonic rudiment of the mouse mammary gland. Dev Biol.1977;61(2):358–365.

27. Kratochwil K, Schwartz P. Tissue interaction in androgen response ofembryonicmammary rudiment ofmouse: identificationof target tissuefor testosterone. Proc Natl Acad Sci USA. 1976;73(11):4041–4044.

28. Axelstad M, Christiansen S, Boberg J, Scholze M, Jacobsen PR,Isling LK, Kortenkamp A, Hass U. Mixtures of endocrine-disrupting contaminants induce adverse developmental effects inpreweaning rats. Reproduction. 2014;147(4):489–501.

29. Yeh S, Hu Y-C, Wang P-H, Xie C, Xu Q, Tsai M-Y, Dong Z,WangR-S, Lee T-H, Chang C. Abnormal mammary gland developmentand growth retardation in female mice andMCF7 breast cancer cellslacking androgen receptor. J Exp Med. 2003;198(12):1899–1908.

30. Soto AM, Brisken C, Schaeberle C, Sonnenschein C. Does cancerstart in the womb? Altered mammary gland development andpredisposition to breast cancer due to in utero exposure to endo-crine disruptors. J Mammary Gland Biol Neoplasia. 2013;18(2):199–208.

31. Plante I, Stewart MK, Laird DW. Evaluation of mammary glanddevelopment and function in mouse models. J Vis Exp. 2011; (53):2828.

32. Suvorov A, Vandenberg LN. To cull or not to cull? Considerationsfor studies of endocrine-disrupting chemicals. Endocrinology.2016;157(7):2586–2594.

33. Markey CM, Luque EH,Munoz De ToroM, Sonnenschein C, SotoAM. In utero exposure to bisphenol A alters the development andtissue organization of the mouse mammary gland. Biol Reprod.2001;65(4):1215–1223.

34. Vandenberg LN, Maffini MV, Schaeberle CM, Ucci AA, Son-nenschein C, Rubin BS, Soto AM. Perinatal exposure to the xen-oestrogen bisphenol-A induces mammary intraductal hyperplasiasin adult CD-1 mice. Reprod Toxicol. 2008;26(3-4):210–219.

35. Haseman JK, Bailer AJ, Kodell RL, Morris R, Portier K. Statisticalissues in the analysis of low-dose endocrine disruptor data. ToxicolSci. 2001;61(2):201–210.

36. Soto AM, Sonnenschein C. The two faces of Janus: sex steroids asmediators of both cell proliferation and cell death. J Natl CancerInst. 2001;93(22):1673–1675.

37. YinW,Maguire SM, PhamB,GarciaAN,DangNV, Liang J,WolfeA, Hofmann HA, Gore AC. Testing the critical window hypothesisof timing and duration of estradiol treatment on hypothalamic genenetworks in reproductively mature and aging female rats. Endo-crinology. 2015;156(8):2918–2933.

38. Kassotis CD, Tillitt DE, LinCH,McElroy JA,Nagel SC. Endocrine-disrupting chemicals and oil and natural gas operations: potentialenvironmental contamination and recommendations to assesscomplex environmental mixtures. Environ Health Perspect. 2016;124(3):256–264.

39. Wasner G, Hennermann I, Kratochwil K. Ontogeny of mesen-chymal androgen receptors in the embryonic mouse mammarygland. Endocrinology. 1983;113(5):1771–1780.

40. Skarda J. Bioassay of steroid hormone agonist and antagonistactivities of anti-androgens on mammary gland, seminal vesiclesand spleen of male mice. J Vet Med A Physiol Pathol Clin Med.2003;50(4):204–212.

41. Boberg J, Christiansen S, Axelstad M, Kledal TS, Vinggaard AM,Dalgaard M, Nellemann C, Hass U. Reproductive and behavioraleffects of diisononyl phthalate (DINP) in perinatally exposed rats.Reprod Toxicol. 2011;31(2):200–209.

42. Christiansen S, Boberg J, AxelstadM,DalgaardM,VinggaardAM,Metzdorff SB, Hass U. Low-dose perinatal exposure to di(2-ethylhexyl) phthalate induces anti-androgenic effects in malerats. Reprod Toxicol. 2010;30(2):313–321.

43. Mandrup K, Boberg J, Isling LK, Christiansen S, Hass U. Low-doseeffects of bisphenol A on mammary gland development in rats.Andrology. 2016;4(4):673–683.

44. Markey CM, Coombs MA, Sonnenschein C, Soto AM. Mamma-lian development in a changing environment: exposure to endocrinedisruptors reveals the developmental plasticity of steroid-hormonetarget organs. Evol Dev. 2003;5(1):67–75.

45. McNally S, Martin F. Molecular regulators of pubertal mammarygland development. Ann Med. 2011;43(3):212–234.

46. Murray TJ, Maffini MV, Ucci AA, Sonnenschein C, Soto AM.Induction of mammary gland ductal hyperplasias and carcinomain situ following fetal bisphenol A exposure.ReprodToxicol. 2007;23(3):383–390.

47. DurandoM, Kass L, Piva J, Sonnenschein C, Soto AM, Luque EH,Mu~noz-de-Toro M. Prenatal bisphenol A exposure induces pre-neoplastic lesions in the mammary gland in Wistar rats. EnvironHealth Perspect. 2007;115(1):80–86.

48. Paine IS, Lewis MT. The terminal end bud: the little engine thatcould. J Mammary Gland Biol Neoplasia. 2017;22(2):93–108.

49. Hinck L, Silberstein GB. Key stages in mammary gland develop-ment: the mammary end bud as a motile organ. Breast Cancer Res.2005;7(6):245–251.

50. Hermes GL, McClintock MK. Isolation and the timing of mam-mary gland development, gonadarche, and ovarian senescence:implications for mammary tumor burden. Dev Psychobiol. 2008;50(4):353–360.

51. Macon MB, Fenton SE. Endocrine disruptors and the breast: earlylife effects and later life disease. J Mammary Gland Biol Neoplasia.2013;18(1):43–61.

52. Mu~noz-de-ToroM,Markey CM,Wadia PR, Luque EH, Rubin BS,Sonnenschein C, Soto AM. Perinatal exposure to bisphenol-A altersperipubertal mammary gland development inmice.Endocrinology.2005;146(9):4138–4147.

53. Wadia PR, Vandenberg LN, Schaeberle CM,Rubin BS, SonnenscheinC, Soto AM. Perinatal bisphenol A exposure increases estrogensensitivity of the mammary gland in diverse mouse strains. EnvironHealth Perspect. 2007;115(4):592–598.



54. Moral R, Wang R, Russo IH, Lamartiniere CA, Pereira J, Russo J.Effect of prenatal exposure to the endocrine disruptor bisphenol Aon mammary gland morphology and gene expression signature.J Endocrinol. 2008;196(1):101–112.

55. Bresciani F. Effect of ovarian hormones on duration of DNAsynthesis in cells of the C3Hmouse mammary gland. Exp Cell Res.1965;38(1):13–32.

56. Williams JM, Daniel CW. Mammary ductal elongation: differen-tiation of myoepithelium and basal lamina during branchingmorphogenesis. Dev Biol. 1983;97(2):274–290.

57. Tucker DK, Macon MB, Strynar MJ, Dagnino S, Andersen E,Fenton SE. Themammary gland is a sensitive pubertal target in CD-1 and C57Bl/6 mice following perinatal perfluorooctanoic acid(PFOA) exposure. Reprod Toxicol. 2015;54:26–36.

58. Hass U, Boberg J, Christiansen S, Jacobsen PR, Vinggaard AM,Taxvig C, Poulsen ME, Herrmann SS, Jensen BH, Petersen A,Clemmensen LH, Axelstad M. Adverse effects on sexual develop-ment in rat offspring after low dose exposure to a mixture of en-docrine disrupting pesticides.ReprodToxicol. 2012;34(2):261–274.

59. Bu~nay J, Larriba E, Moreno RD, Del Mazo J. Chronic low-doseexposure to a mixture of environmental endocrine disruptors in-duces microRNAs/isomiRs deregulation in mouse concomitantwith intratesticular estradiol reduction. Sci Rep. 2017;7(1):3373.

60. SobolewskiM,ConradK, Allen JL,WestonH,Martin K, LawrenceBP, Cory-Slechta DA. Sex-specific enhanced behavioral toxicityinduced by maternal exposure to a mixture of low dose endocrine-disrupting chemicals. Neurotoxicology. 2014;45:121–130.

61. Rajapakse N, Silva E, Kortenkamp A. Combining xenoestrogens atlevels below individual no-observed-effect concentrations dra-matically enhances steroid hormone action. Environ Health Per-spect. 2002;110(9):917–921.

62. Rajapakse N, Silva E, Scholze M, Kortenkamp A. Deviation fromadditivity with estrogenic mixtures containing 4-nonylphenol and4-tert-octylphenol detected in the E-SCREEN assay. Environ SciTechnol. 2004;38(23):6343–6352.

63. Silva E, Rajapakse N, Kortenkamp A. Something from “noth-ing”—eight weak estrogenic chemicals combined at concentrationsbelow NOECs produce significant mixture effects. Environ SciTechnol. 2002;36(8):1751–1756.

64. DefargeN, Takacs E, LozanoVL,MesnageR, Spiroux deVendomoisJ, Seralini GE, Szekacs A. Co-formulants in glyphosate-based her-bicides disrupt aromatase activity in human cells below toxic levels.Int J Environ Res Public Health. 2016;13(3):13.

65. Kortenkamp A. Low dose mixture effects of endocrine disrupters:implications for risk assessment and epidemiology. Int J Androl.2008;31(2):233–240.

66. Czolowski ED, Santoro RL, Srebotnjak T, Shonkoff SBC. Towardconsistent methodology to quantify populations in proximity to oiland gas development: a national spatial analysis and review. En-viron Health Perspect. 2017;125(8):086004.

67. Balise VD, Meng CX, Cornelius-Green JN, Kassotis CD, KennedyR, Nagel SC. Systematic review of the association between oil andnatural gas extraction processes and human reproduction. FertilSteril. 2016;106(4):795–819.



Documents

Prenatal Exposure to Unconventional Oil and Gas Operation ......Prenatal Exposure to Unconventional Oil and Gas Operation Chemical Mixtures Altered Mammary Gland Development in Adult